Supplementary information

Materials and Methods

Patient samples, cell purification and sorting

55 samples from CLL patients were obtained after written informed consent in accordance with local institutional guidelines and the Declaration of Helsinki.The main clinical and biological characteristics of the patient cohort are summarized in Table 1. The expression of CD38 was checked by flow-cytometry. B cells were purified from PBMC by negative selection.20 The intraclonal CD38- component of CLL samples with bimodal CD38 expression was obtained by incubation with anti-CD38 monoclonal antibodies (mAbs, produced in-house) and magnetic bead separation (Life Technologies, Monza, Italy). Alternatively, CLL cells were triple-labeled with anti-CD5-FITC, -CD38-PE and -CD19-APC (eBioscience, Hatfield, UK) and physically sorted by high-speed cell sorting (10,000-30,000 cells/s) using a MoFlo High Speed Cell Sorter (Beckman Coulter, High Wycombe, UK).

Gelatin zymography assays, degradation assays and confocal microscopy

For constitutive expression of MMP-9, CLL cells (2x106) were lysed (PBS + 2% Triton-X100 + 0.1% SDS + 5mM EDTA and protease inhibitors) and lysates analyzed for gelatinolytic activity on 8% polyacrylamide gels containing 1 mg/ml gelatin (Sigma, Milan, Italy). After electrophoresis, gels were rinsed 3 x 10 min in 2.5% Triton X-100, followed by overnight incubation in Zymogram Buffer (50 mM Tris pH 7.5, 0.2 M NaCl, 5mM CaCl2,0.05% Brij35) at 37°C. Gels were stained with 0.25% Coomassie blue (Sigma) and areas of gelatinolytic activity visualized as transparent bands. Images were acquired using the ImageQuant Las4000 gel imager: bands were quantified using the ImageQuantTL 7.0 software (GE Healthcare, Milan, Italy).Basal MMP-9 expression by cells was expressed as arbitrary units and standardized over protein content. The same lysate from a highly positive CLL patient was used as an internal loading control in different gels, constantly yielding a band of comparable intensity. For inducible MMP-9 expression, CLL cells (2x106) were cultured in RPMI+0.1% FCSin the presence of fibronectin (FN 10 mg/ml, Sigma), CXCL12 (150 ng/ml, R&D Systems, Space, Milan, Italy) and anti-CD38 agonistic mAb (IB4, 10 g/ml, produced in-house) for 24h and the conditioned medium concentrated 10-fold using Amicon Ultra 10k (Millipore, Vimodrone, Milan, Italy). In selected experiments, cells were pre-treated (1h) with wortmannin (PI-3K inhibitor, 30 nM) or UO126(ERK1/2 inhibitor, 5 μM) before the 24h incubation.MMP-9 secretion is represented as a ratio between the different conditions. The MMP-9 activity of cells cultured in 0.5% BSA is considered as a negative control.

Matrix degradation experiments were performed using glass coverslips coated overnight with 100 μg/ml FITC-labeled gelatin in PBS, and fixed with 0.5% glutaraldehyde. After washing with PBS, poly-lysine was added to coverslips (30 min, 37 °C), washed with PBS and medium, and CLL cells in RPMI+0.1% FCS added and incubated for 24 h. Cells were fixed with PFA 4% and stained with CytoRed (Molecular Probes, Life Technologies). In indicated experiments, the agonistic anti-CD38 mAb was added to cells and cells pre-treated with specific inhibitors, as described above. Coverslip were then analyzed using a TCS SP5 laser scanning confocal microscope with 4 lasers (Leica Microsystems, Milan, Italy) running the LAS AF Version Lite 2.4 software (Leica Microsystems) and processed with Photoshop (Adobe Systems, San Jose, CA).

For co-localization experiments, glass coverslips were coated with rhVCAM-1 (R&D, Milan Italy,overnight at 4 °C) andCLL cells incubated (1h, 37 °C) before fixation and permeabilizaton. The Abs used for staining were: anti-MMP-9 (Bio-Optica, Milan, Italy), anti-P-Tyr (BD Biosciences, Buccinasco, Italy), biotin-conjugated anti-CD38 (produced in-house), biotin-conjugated anti-CXCR4 (R&D), Alexa647-conjugated anti-CD49d (AbD Serotec, Puchheim, Germany), anti-CD38 (AbD Serotec), Alexa-568-conjugated avidin (Life Technologies) and Alexa488-conjugated anti-mouse-IgG (Life Technologies).Images were acquired using a TCS SP5 laser scanning confocal microscope and processed as above.

Enzyme-linked immunosorbent assay

Plasma from CLL patients were collected after two centrifugal steps for a complete platelet removal. Samples were then stored at -80 °C. MMP-9 levels were measured with a quantitative enzyme-linked immunosorbent assay (ELISA) following manufacturer’s instruction (Human MMP-9 Immunoassay, Quantikine® ELISA, R&D Systems).

Construct, lentiviruses and infection

Lentiviral particles containing the genetic material for CD38 or for GFP were generated as described17 and designated S38W and SEW, respectively. Freshly purified or thawed PBMC preparations from CLL patients were plated (106/ml) in RPMI-1640 medium + 10% fetal calf serum (FCS, Seromed, Berlin, Germany) in the presence of viral particles (48 h, 37 °C) titrated to obtain CD38 expression levels commonly observed in CLL patients.21

Western blotting

After incubation, CLL cells were lysed and western blot analysis performed as indicated.20 The Abs used were: anti-p-Erk1/2, anti-Erk1/2 (both from BD Biosciences), anti-p-p38 (Cell Signaling Technology, Danvers, MA), anti-p-Akt (Santa Cruz Biotechnology, Magenta, Italy). HRP-conjugated anti-mouse-IgG (Perkin Elmer, Monza, Italy) or anti-rabbit (Santa Cruz Biotechnology) abs were used as secondary reagents.

RT-PCR analysis of MMP-9 transcription

Total RNA was extracted from CLL lymphocytes using RNeasy spin columns (Qiagen, Crawley, UK) and reverse-transcribed using a Clontech first-strand cDNA synthesis-kit (Clontech, Between Towns, UK). RT-PCRwas performed using MMP-9-specific primers 5-CATCGTCATCCAGTTTGGTG-3 (forward) and 5-GCCTTGGAAGATGAATGGAA-3 (reverse) and RPS14 (house-keeping gene) primers 5-GGCAGACCGAGATGAACTCT-3 (forward) and 5-CCAGGTCCAGGGGTCTTGGT-3 (reverse). The RPS14:MMP-9 transcription ratio was calculated from the crossing points of each gene and the relative expression of MMP-9 was calculated using the comparative crossing point method (ct).

Statistical analyses

All statistical tests were performed using the Graphpad Prism 5.0 software (Graphpad Software, San Diego, CA, USA).

Supplementary Tables and Figures

Supplementary Figure 1. (A-B) Expression and activity of MMP-9 measured by degradation assay using gelatin-FITC coated coverslips confirmed the different propensity of CD38+ and CD38- cells (n=6) to digest the extracellular matrix (p<0.0001). (C) When considered as continuous variables, a positive correlation between CD38 and MMP-9 expression was highlighted (R2= 0.16, p= 0.002).